AudVis - audio visualization add-on

Hello Blender

There is a new version of Blender. This is mainly a bugfix version, so version number changed only to 1.9.1.

  • Shape modifier for grease pencil - this stopped working after adding or removing strokes in frame 0. Fixed.
  • Video - (a feature which captures image from your camera, useful with Party Mode + Realtime Analyzer) - blender crashed cca after 10 seconds. Fixed.
  • Shaper modifier - added UV Map and Vertex Color for Mesh objects.
  • Shape modifier - grease pencil - added Layer filter
  • Shape modifier - added button “Clean Data” - this cleans the baked data and removes target shape modifier, uv map, vertex color layer… They will be regenerated when needed.
  • Video - added option Temp path - you can choose where AudVis saves temporary images
  • Shape modifier and Spectrogram optimized - now they are a bit faster

Last thing, I have unpublished gumroad page, so now AudVis is available only on Blender Market.

If you have any questions, I’m here for you. Enjoy AudVis!

1 Like

Is it possible to use multiple audio files from sequencer. So I could choose tha some models react to different audio files

1 Like

Good question. For now it’s not possible, but this will be easy to implement. I will take a look at that.
Do you want to use it with Shape modifier?

For now I’d like to use it with sequence analyzer. Haven’t look into the other features yet.

By the way this addon is god damn awesome! You are making the great Blender much greater!

Thank you very much :sunglasses:

I have sent you a message. When you confirm the proposed solution is OK, it will be a part of the next release :slight_smile:

(I get this kinda error, not sure if its about the addon.

When I render image the console says:

Detected 4 dependency cycles
tekkno1.mp3
tekkno1.mp3
tekkno1.mp3
tekkno1.mp3
tekkno1.mp3

After that it renders. But it takes pretty long to start rendering. Any Idea? )

Now I am not experiencing that anymore.

Do you think that the method you are using can be used for making an audio monitor for the VSE too?

@tin2tin Yes, in AudVis it’s called “Sequence Analyzer”. Here is a bit outdated video, but you can still use it as a reference (start from 2:30)

@vihrea Detected 4 dependency cycles - sometimes I see this error, but I don’t think it’s caused by AudVis. It’s probably just some internal data problem. When I saw it, nothing was broken, it was just a warning.

1 Like

how are you getting audio from the sequencer? are you analyzing a single strip, or are you getting a mix somehow?

Short answer: it’s a sum of all sound sequences and realtime analyzer if enabled.
Long answer:

  • First step is done after changing a frame. Here I’m getting last 0.1 second of each sound. You can set this with the field “Sample Size” and “Audio Offset”, but it’s also affected if you check “Use Better Filters” option.
  • If “Use Better Filters” is checked, there is happening also some more magic, using Audaspace (for more detailed info, read https://docs.blender.org/api/current/aud.html)
  • Then I’m doing Fast Fourier Transform on these sound samples and store the result, each one separately. If you have input with 44800 Hz, max. available frequency after FFT is 22400 Hz.
  • When you call a driver, for example audvis(0, 100), audvis makes this with each of these FFT results:
    • get frequencies 0-100
    • get average value from the step above
  • sum up the values from previous step (realtime + sequence 1 + sequence 2 + sequence 3…)
  • make all the magic things (add noise, add, max, factor…)
  • return value

Interesting, i had considered doing something similar to attempt to implement a vu meter in the sequencer, but I had thought it would be too complicated and too slow to be usable, and had resigned myself to waiting for someone to implement it in C code.
I’d love to take a look at your code and maybe do some speed tests and see what its capable of.

answered in personal message

This addon is high quality :+1:

I supported via blendermarket. I used the addon in this project:

Forum topic

1 Like

Yanno, just when i thought Blender couldn’t get any cooler, along comes this plugin! I can’t wait to start playing with this! I’ve got so many ideas …

Thank you so much for your work

1 Like

Hi there,

I’m back to introduce you a new version of AudVis, 2.0.0.

What’s new:

  • Added support for multiple channels
    Screenshot from 2020-02-21 20-05-33
  • Added support for selectively responding to different sequences
    Screenshot from 2020-02-21 20-08-32
  • Improved speed of Driver Values -> Use Better Filters -> Use Curve - now it’s not slowing anything
  • Added new functionality - recording the sound. It’s under Real Time Analyzer.Screenshot from 2020-02-21 20-01-08
  • Removed fake modifier
  • Driver Values presets
    Screenshot from 2020-02-21 20-10-33

Enjoy!

1 Like